Excavation Reinforcement Learning Using Geometric Representation
This work addresses a domain-specific problem in robotics for excavation tasks, presenting an incremental improvement by introducing a representation learning approach to enhance RL efficiency and transferability.
The paper tackles the challenge of excavating irregular rigid objects in clutter by using reinforcement learning to plan sequences of excavation trajectories from point clouds, and shows that a separately learned geometric representation reduces training time while achieving similar performance to end-to-end RL, with the representation-based policy outperforming end-to-end RL when transferred to real scenes.
Excavation of irregular rigid objects in clutter, such as fragmented rocks and wood blocks, is very challenging due to their complex interaction dynamics and highly variable geometries. In this paper, we adopt reinforcement learning (RL) to tackle this challenge and learn policies to plan for a sequence of excavation trajectories for irregular rigid objects, given point clouds of excavation scenes. Moreover, we separately learn a compact representation of the point cloud on geometric tasks that do not require human labeling. We show that using the representation reduces training time for RL, while achieving similar asymptotic performance compare to an end-to-end RL algorithm. When using a policy trained in simulation directly on a real scene, we show that the policy trained with the representation outperforms end-to-end RL. To our best knowledge, this paper presents the first application of RL to plan a sequence of excavation trajectories of irregular rigid objects in clutter.